Official Helium Community - hip-55-validator-challenges

# HIP 55: Validator Challenges - Author(s): @professor , @hashc0de, @mack - Start Date: 2022-02-03 - Category: Technical - Original HIP PR: https://github.com/helium/HIP/pull/359 - Tracking Issue: https://github.com/helium/HIP/issues/362 - Status: In Discussion # Summary [summary]: #summary This HIP proposes a change to how Proof-of-Coverage (PoC) Challenges are generated and submitted to the Helium blockchain to allow for further network scalability and to lower the hardware complexity/cost of Hotspots. Specifically, it moves the responsibility of PoC Challenge creation to Validators and consequently proposes moving the economic reward for creating Challenges to this group as well. # Rendered view https://github.com/helium/HIP/blob/main/0055-validator-challenges.md (edited)

Add HIP for validator challenges by Vagabond · Pull Request #359 · ...

Rendered

HIP 55: Validator Challenges · Issue #362 · helium/HIP

Author(s): @Vagabond, @andymck, @abhay Start Date: 2022-02-03 Category: Technical Original HIP PR: #359 Tracking Issue: #362 (this) Status: In Discussion Summary This HIP proposes a change to how P...

HIP/0055-validator-challenges.md at main · helium/HIP

Helium Improvement Proposals. Contribute to helium/HIP development by creating an account on GitHub.

so If I'm reading this correctly it would be moving 0.9% more of the rewards to the validators?

1

BadGuyTy

so If I'm reading this correctly it would be moving 0.9% more of the rewards to the validators?

Yes for doing the challenge creation portion of the PoC (edited)

Peso die Schildkröte 🇩🇪 pinned a message to this channel. 02/14/2022 7:40 AM

I have no problems with the validators doing the challenges but I do have issues with the further reduction of possible rewards to hotspot owners. We already lost the consensus rewards possibilities and now challenges?

BadGuyTy

I have no problems with the validators doing the challenges but I do have issues with the further reduction of possible rewards to hotspot owners. We already lost the consensus rewards possibilities and now challenges?

Just take a look how much the last challenge creation rewarded you. This is a small price for not having to deal with relayed hotspots anymore (edited)

3 cents is 3 cents (didn't actually check that value) (edited)

So how much would we loose with this hip

In my opinion all work done needs to be rewarded. You can't move work over to validators and expect them to do it for free. If that was the case they'd simply block those ports and have fun not getting any PoC... but that could be too cynical

1

secretboy

So how much would we loose with this hip

Approx this 3x per day (edited)

Peso die Schildkröte 🇩🇪

Approx this 3x per day (edited)

So no more of those rewards if i understood you correctly?

Peso die Schildkröte 🇩🇪

Approx this 3x per day (edited)

it's actually x 1.5 as only 50% of challenges are accepted currently

1

secretboy

So no more of those rewards if i understood you correctly?

Yep

0.9% split across hotspots who have created challenges within an epoch. today that's roughly 30 blocks so about 30k hotspots today. that # goes up as the number of hotspots on chain goes up.

Peso die Schildkröte 🇩🇪

Yep

Eh those are the lowest rewards but what do we benefit from this?

it's all about tradeoffs. - no relayed hotspots / port forwarding - scalability to millions of hotspots - no syncing - reduced sd card write wear issues - better chance of actually delivering your witness receipts so potentially more participation in poc downside - but my ubi (universal basic income, for those not familiar with the acronym) (edited)

5

secretboy

Eh those are the lowest rewards but what do we benefit from this?

Network stability and no more relayed hotspots (edited)

1

So what bout the rewards will we see a increase on that?

This will result in my lone wolf getting 0 rewards and I'm for this HIP.. but I get it rewards are rewards.

2

secretboy

So what bout the rewards will we see a increase on that?

Rn, half the PoC is going through, I hope we can expect 90% or close to 95 after this HIP (edited)

Peso die Schildkröte 🇩🇪

Rn, half the PoC is going through, I hope we can expect 90% or close to 95 after this HIP (edited)

So more rewards then?

Also for what it's worth, it's not likely to increase people rewards. Just an overall better and healthier network. Because if everyone is getting and seeing more PoC the rewards will be split into smaller pieces since the size of the pie isn't changing. Now, would this mean you should see more stable rewards, yes. More rewards, not necessarily

secretboy

So what bout the rewards will we see a increase on that?

yes and no and depends. yes if your hotspot is relayed, no you will lose the challenge rewards, depends if you are not relayed than you will also lose rewards as the relayed hs will be getting more

secretboy

So more rewards then?

Well, thats not how it works. But if we simplify alot, yes

hashc0de

it's all about tradeoffs. - no relayed hotspots / port forwarding - scalability to millions of hotspots - no syncing - reduced sd card write wear issues - better chance of actually delivering your witness receipts so potentially more participation in poc downside - but my ubi (universal basic income, for those not familiar with the acronym) (edited)

Also no syncing, sd card issues, port forwarding, etc

1

Eh decent just hope that i dont loose more with this hip

so isn't this the same as light hotspots?

@hashc0de was there consideration given to having the CG assign other online validators to perform the PoC work? like when proposing a list of hotspot keys, also including the a randomly selected online validator to perform the challenge work? that way, the rest of the validator group is doing something rather than sitting idle and just adding more work to the same validators that are in CG?

3

BadGuyTy

so isn't this the same as light hotspots?

It’s one of the components needed for light hotspots to function

Yea, more plug and play. No hidden fees. Jim Carrey (cable guy) would be upset

4

PaulM

@hashc0de was there consideration given to having the CG assign other online validators to perform the PoC work? like when proposing a list of hotspot keys, also including the a randomly selected online validator to perform the challenge work? that way, the rest of the validator group is doing something rather than sitting idle and just adding more work to the same validators that are in CG?

have a look at the HIP. actually mentioned in there

07:52

it's definitely an idea we like. increased engineering scope at this time though.

I can also imagine this will massively reduce chain bloat by not giving dust to hundreds of thousands of hotspots in reward transactions

hashc0de

have a look at the HIP. actually mentioned in there

dang. i need to read better

rawrmaan

I can also imagine this will massively reduce chain bloat by not giving dust to hundreds of thousands of hotspots in reward transactions

hopefully before we reach a million

1

rawrmaan

I can also imagine this will massively reduce chain bloat by not giving dust to hundreds of thousands of hotspots in reward transactions

it's not immaterial but not an explicit problem today. 10s of thousands per epoch.

So if network is more stable will we see a increase in witnesses and beacons because they were lowered earlier this month?

secretboy

So if network is more stable will we see a increase in witnesses and beacons because they were lowered earlier this month?

if you have a look at the hip, it specifies that there will only be a fixed set of challenges per block. and that can be adjusted if validators are able to handle more transactions per block. TBD.

secretboy

So if network is more stable will we see a increase in witnesses and beacons because they were lowered earlier this month?

I’d say it’s more like: more hotspots will beacon and witness consistently. Right now beaconing is highly dependent on how well the p2p network works, and witnessing fails about 30-50% of the time due to p2p issues

but also ^

capcom

I’d say it’s more like: more hotspots will beacon and witness consistently. Right now beaconing is highly dependent on how well the p2p network works, and witnessing fails about 30-50% of the time due to p2p issues

So we can expect a little increase?

No clue. But things should work as they should and be less inconsistent

PaulM

dang. i need to read better

Quick question on this line of thought, what about the validators in CG? Would they also need to perform challenge creations while in to the hotspots they are connected to? I think so, just curious on what it'd mean for the group. P.S - I tagged you Paul because if you've seen some "adverse affects" with connections and CG already (edited)

Yeah, that was my line of thinking - trying to minimize the incremental work and incoming connections for the CG

If this is something required for light hotspots is this something that is currently being tested on the testnet?

Anthonyra

Quick question on this line of thought, what about the validators in CG? Would they also need to perform challenge creations while in to the hotspots they are connected to? I think so, just curious on what it'd mean for the group. P.S - I tagged you Paul because if you've seen some "adverse affects" with connections and CG already (edited)

recall the 2 roles in the hip: 1. CG members create ephemeral keys and write to the chain 2. All validators serve as gateway services (read-only RPC) Group 1 will be also responsible for creating receipts transactions, receiving witness and challengee receipts (edited)

crazyhonybadger

If this is something required for light hotspots is this something that is currently being tested on the testnet?

Yup see #announcements

1

we have real hardware doing this on the testnet as of friday i think. we've been using virtual hotspots for a few weeks now

07:59

HIPs come with code.

If I'm reading correctly, all hotspots in an H3 that is selected for PoC will send a request to the Challenge Validator (possible still a CG member) asking is they are the target? is this going to flood the CG with requests?

hashc0de

recall the 2 roles in the hip: 1. CG members create ephemeral keys and write to the chain 2. All validators serve as gateway services (read-only RPC) Group 1 will be also responsible for creating receipts transactions, receiving witness and challengee receipts (edited)

I guess, I shouldn't ask it as a question but more as a concern. Because it does say whether they are in consensus or not will create challenges based on the blocks metadata and challenge those hotspots that are clients of it. I'm curious if CG should be excluded from creating challenges and collecting receipts while in?

Anthonyra

I guess, I shouldn't ask it as a question but more as a concern. Because it does say whether they are in consensus or not will create challenges based on the blocks metadata and challenge those hotspots that are clients of it. I'm curious if CG should be excluded from creating challenges and collecting receipts while in?

Specifically it is the CG that creates ephemeral keys. The interesting part of the design is that it can just be added as block metadata and doesn't need to be part of the transaction budget per block.

Correct, and maybe I'm not explaining myself well enough. But, like Paul said above. All validators will see that block and based on their client list ping the hotspots in the H3 hex that's a target. It's mentioned that hotspots could be randomly distributed or delegated to a specific validator (which is a good idea in my opinion) but we see connection issues already in CG with PoC receipts and the like. If that validator has a lot of clients in the hex would it cause issues?

depending on how long the poc_timeout is, however, that CG member may no longer be in consensus and still be serving as the Challenger.

we are talking about targeting at H3 res5, right? that could be a thousand hotspots, multiplied by however many challenges generated per Challenge Validator? sounds like a lot of requests being sent to CG members

08:15

also, what incentive is there for validators to run the gateway service when not in CG?

none with the implementation we have so far. but attestations could be used as a metric for validator affinity / delegation in the future.

08:29

since all lookup responses are signed, they can be used to penalize lying validators and potentially bless useful validators. we haven't scoped it yet but open to exploration (penalty adjustments, election weighting, etc.) (edited)

I'm not sure I fully understand this. Could I just keep the port closed and ignore the extra work?

you could!

What's the incentive not to do that?

potential rewards (when creating challenge keys while in the CG) (edited)

is there an estimated time for release?

08:34

when will regular hotspots become light hotspots?

08:35

I think the last timeline was Q4 2021

08:35

This HIP is part of the process. Latest roadmap for LH will always be here: https://docs.helium.com/mine-hnt/light-hotspots

It sounds to me, like some of us would actually get more rewards due to this HIP, when you factor in down time lost to syncing. Since we have containerized miners now, it might not be as big of an impact. But I'll say I've lost many days over the last few months due to syncing, so it sounds like a win overall from my perspective.

manicadam

It sounds to me, like some of us would actually get more rewards due to this HIP, when you factor in down time lost to syncing. Since we have containerized miners now, it might not be as big of an impact. But I'll say I've lost many days over the last few months due to syncing, so it sounds like a win overall from my perspective.

Not if everyone else having the same issues as you now don't have those issues. It'll result in more stable and predictable rewards though

getting to "stable and predictable" is 100% a goal of this hip.

2

3

what is meant by "the target validator(s) [to which a hotspots are connected via a durable gRPC connection] could be random or specified"? the hotspot owner could pick the validator to connect to? i think random might be better (wondering if a security question could arise if a source builder pointed hotspots to their own validators to rubber stamp)

It can be configured. would love to understand your security concerns around a colluding validator, though. The validator creating the challenge is not necessarily the one that is providing read access to blockchain data. All responses from the latter validator are signed / attested by the validator as well. (edited)

will do my best here. just trying to figure out how to think about mitigating gaming the system. from your response, this concept may be impractical. nonetheless, my thought was that if you have a validator (or, as it sounds like, multiple validators) and hotspot(s) acting in concert, you could generate challenges and then approve the results from your favored hotspots. all other hotspot PoC challenges would be handled normally, but if you control the source validator (and it happens to select one of your hotspots for a challenge), you could approve it regardless of actual coverage. still trying to wrap my head around H3Dex and HIP 55, so this may be impossible. or irrelevantly-infrequent.

09:39

risk would be the slashing concept described. may also need control of a geographical area without honest hotspots or otherwise risk contradictory assertions.

firsthummus

will do my best here. just trying to figure out how to think about mitigating gaming the system. from your response, this concept may be impractical. nonetheless, my thought was that if you have a validator (or, as it sounds like, multiple validators) and hotspot(s) acting in concert, you could generate challenges and then approve the results from your favored hotspots. all other hotspot PoC challenges would be handled normally, but if you control the source validator (and it happens to select one of your hotspots for a challenge), you could approve it regardless of actual coverage. still trying to wrap my head around H3Dex and HIP 55, so this may be impossible. or irrelevantly-infrequent.

I think that's the right line of thinking. But some additional thoughts: 1. You need to be in consensus to start generating ephemeral keys. those keys are tied to you as a validator. that key (public key + block hash) generates an h3 zone and you select a particular hotspot in that zone. only you as the validator know which hotspot in that zone is being challenged (but it must be within that zone). 2. Any validator can provide data on chain about a h3 zone that's current being challenged. they use the emphemeral public key + block hash to generate a random h3 zone for selection like the challenging validator does. 3. all hotspots within the h3 zone are told by their favorite validator (configuration, local list of "safe" validators, random selection, etc.) that they could potentially be challenged and are provided the routing (pubkey, ip address, port, etc.) of the challenging validator so the risks you mentioned i think boil down to this as a hotspot owner who wants to game the system: - you need to have validators in consensus to generate keys and only generate keys in the zone your hotspots live. - you potentially need to have have colluding validators who tell you your zone is being challenged. - your colluding cg validator needs to return an onion to you

1

I've got an overall concern that I guess is more about light hotspots and long-living gRPC connections rather than this HIP specifically, but I guess I'll ask here (let me know if I should move this somewhere else): I think the dev team and validator operators need to carefully think thru the problems around having many simultaneous active connections. We've run into kernel and system limits in the past and mostly worked through them as we found them, but if we expect validators to have 1-2 active connections, we should think about other systems that have been tuned to that kind of workload and do that preemptively. (edited)

There's also the spam issue. I know Solana is working on a switch from UDP to QUIC right now due to spammers being able to easily spoof IPs. These are definitely "high-scale" problems but there's no question that validators will need to be able to defend themselves from DoS-like activity

15:02

I guess it's pretty easy to have the gateway sign at the beginning of a connection and block any IPs that don't have a signature from a known gateway/limit the amount of new connections per gateway per time period

Since the requests are over gRPC, and not a custom protocol over libp2p, we can take advantage of the 25+ years of scaling web servers, caching, rate limiting, etc.

rawrmaan

There's also the spam issue. I know Solana is working on a switch from UDP to QUIC right now due to spammers being able to easily spoof IPs. These are definitely "high-scale" problems but there's no question that validators will need to be able to defend themselves from DoS-like activity

QUIC is already multiplexed UDP. It's being called HTTP/3 because it uses the HTTP/2 (gRPC uses this) style multiplexing and some congestion control.

15:56

HTTP/2 is interesting here because it specifically is robust to long lived connections.

hashc0de

Since the requests are over gRPC, and not a custom protocol over libp2p, we can take advantage of the 25+ years of scaling web servers, caching, rate limiting, etc.

Totally agree, it's not breaking new ground, but I wanted to call it out because so many validator operators appear to be running stock OS with docker or from source and aren't thinking about tuning for things like "a big webserver" But I'm with you, we shouldn't be inventing new solutions, just giving guidance (or setting defaults in the container, if it matters) for what operators need to tweak based on what works for similar systems.

1

I think this conversation is good though, especially for the itinerant operators who don't really know what they're doing and yoloed a docker container up somewhere.

Competence of validator operators aside, this is adding more load to the CG while there are 3400 other validators that have lots of spare capacity. Why put it only on the CG? I know the answer will be that the other validators will be providing the gateway service to hotspots however there is no incentive for them to do so.

It's a reasonable point. we could certainly add some kind of value to attestations but not sure how that increases the complexity.

farm out the part of the challenge work to the other validators and let them in on the 0.9%, at least its something

how do you select a validator for the challenge work? you know the CG member is alive and active. (edited)

Validators heartbeat every 100 blocks. Have the CG put something in the block meta data selecting them to challenge. Could do it along with a hex.

Forgive me if this is an idiotic question due to my lack of validator knowledge ..

Could there just be a parallel second "Challenger CG" that just takes care of the challenger duties and then passes the info to the current CG to do the rest of the work?

17:57

So a plug-in replacement of what challenger hotspots are doing now?

17:57

We've got plenty of validators, right?

There could, at the cost of added complexity. This proposal (rightfully so) is attempting to balance complexity with time to implement so we aren't waiting forever for light hotspots.

1

PaulM

There could, at the cost of added complexity. This proposal (rightfully so) is attempting to balance complexity with time to implement so we aren't waiting forever for light hotspots.

Yeah, figured there would be a trade off

PaulM

There could, at the cost of added complexity. This proposal (rightfully so) is attempting to balance complexity with time to implement so we aren't waiting forever for light hotspots.

Was hoping that the existing framework that currently is in place could just be adapted.

ElonTusk | 5Gchipped

Forgive me if this is an idiotic question due to my lack of validator knowledge ..

Could there just be a parallel second "Challenger CG" that just takes care of the challenger duties and then passes the info to the current CG to do the rest of the work?

Would an issue with this be since I can't be invited to join the 'validation CG" while I am in the "Challenger CG" it behooves my validator to say no to the "Challenger CG" ? (edited)

If this goes thru, which probably will, when would this take action?

KeithR

Would an issue with this be since I can't be invited to join the 'validation CG" while I am in the "Challenger CG" it behooves my validator to say no to the "Challenger CG" ? (edited)

I would guess there'd be an algorithm that would move you to the VCG and fill in the CCG with the next validator in the queue.

ElonTusk | 5Gchipped

Forgive me if this is an idiotic question due to my lack of validator knowledge ..

Could there just be a parallel second "Challenger CG" that just takes care of the challenger duties and then passes the info to the current CG to do the rest of the work?

This is mentioned in the HIP btw.

hashc0de

Specifically it is the CG that creates ephemeral keys. The interesting part of the design is that it can just be added as block metadata and doesn't need to be part of the transaction budget per block.

Here’s one of the downsides.

20:19

Block metadata is made and stored by the CG along with the block. So if there’s a separate group, we’d have to retain the request txn

hashc0de

This is mentioned in the HIP btw.

Doh! I'll go read it again. Lol

It’s a dense one.

2

20:21

As with all these proposed changes, I worry if people will truly understand the technical complexity, so we’ll try to be more available. AMA on Wednesday as well.

Maybe just me, but some graphic/flow chart could be handy for people to see the flow of tokens, cause/action, etc. :)

Roger.

20:23

Good feedback for sure

20:24

Let me see if I can get Pete to help with that

2

All Hotspots, upon receipt of a challenge notification, will send a request over gRPC to the Challenger Validator to check if they are the target.

20:42

Why does the hotspot ask the validator if they are the target? Why doesn't the validator just issue the challenge to the target in the first place?

hashc0de

This is mentioned in the HIP btw.

Just got to the part you were referring to. Literally exactly what I was talking about. Lol Thanks.

Is HIP55 up for a vote yet?

Not yet

ElonTusk | 5Gchipped

Why does the hotspot ask the validator if they are the target? Why doesn't the validator just issue the challenge to the target in the first place?

The IP address of the Hotspot could be unknown, behind a relay, etc.

2

21:58

We specifically want the hotspot to reach out to the more stable network participant.

NickL

Is HIP55 up for a vote yet?

Champing at the bit? Get a validator on testnet and you can help test now

hashc0de

As with all these proposed changes, I worry if people will truly understand the technical complexity, so we’ll try to be more available. AMA on Wednesday as well.

Will that be here on Discord or in Zoom? I finally figured out how events work on discord... (edited)

Anthonyra

Will that be here on Discord or in Zoom? I finally figured out how events work on discord... (edited)

discord stages AFAIK

06:35

theres an event-stage under voice <<

BFGNeil - Trackpac.io

discord stages AFAIK

I didn't scroll up to see the 3 active events facepalm

06:35

https://discord.gg/helium?event=942824074801979412

2

We specifically want the hotspot to reach out to the more stable network participant.

@ElonTusk | 5Gchipped Re this ^^^. This is helps keep hotspots from needing incoming connections via port forwarding or similar firewall rules.

1

No blockchain sync / no relays / no port forwarding / just plain http (basically) I’m beyond excited about this evolution (edited)

6

So my understanding is Validators will make more by taking care of blockchain sync so no more relayed?

1

FamilyJ

So my understanding is Validators will make more by taking care of blockchain sync so no more relayed?

0.9%

I can't wait for this to happen!!! In the last 5 days I have lost 50% of my beacons and 40% of my witnesses to the p2p black hole! If everyone else is in the same boat, then half the activity of the network isn't captured! Do you think HIP will also remove the restriction on the number of witnesses? Out of the witnesses that were succesfully reported, only 18% were rewarded (11% of the total). I assume the rest lost the lottery to be in the 14 rewarded. I understand the impact to earnings won't be huge (though something is better than nothing) but I'd like to able to see how much coverage I am providing. (edited)

We probably won't be removing the restrictions on witness counts with this HIP but there's a little note in there about "regional challenges". We have this concept in mind where instead of a single Hotspot being challenged, an entire hex is challenged, and everyone in that hex beacons and witnesses at the same time (while staying in compliance of local radio regulations). this would give the entire area some work to do to prove each other and also able to reward the entire area at once for the information the blockchain received. This is just a concept though and needs some thinking.

3

@hashc0de forgive me if this was answered. I started pushing customers onto helium once again. I wasn't very happy last time I did. I've been relatively happy this time. My sensors get their data moved and it's been stable. I appreciate the hard work that you guys have done. I'm all for light hotspots and any way to scale this network. For the people that are using it with paying clients. Is there any chance that switching this over to light hot spots will affect data moving through the network?

The_Professional

@hashc0de forgive me if this was answered. I started pushing customers onto helium once again. I wasn't very happy last time I did. I've been relatively happy this time. My sensors get their data moved and it's been stable. I appreciate the hard work that you guys have done. I'm all for light hotspots and any way to scale this network. For the people that are using it with paying clients. Is there any chance that switching this over to light hot spots will affect data moving through the network?

From my understanding, hotspots will not transmit packets if they're not fully synced, as they don't have the latest view of state channels and OUIs. So, if syncing is no longer necessary and hotspots can transmit data (nearly) 100% of the time, the network will be more reliable.

4

that’s right. There’s also work to get hotspots transferring packets without being fully synced that’s getting soak time. Doing this will allow more consistent and stable coverage for devices, and less coverage downtime

6

5

Couldn’t have said it better than these two. #teamnosyncneeded

Thank you @rawrmaan @hashc0de @cokes for the response. As long as data keeps moving through the network I don't care. It's got my vote. Carry on ladies and gentlemen.. sorry Cokes. (edited)

… and woman! (One of us at least)

hashc0de

We probably won't be removing the restrictions on witness counts with this HIP but there's a little note in there about "regional challenges". We have this concept in mind where instead of a single Hotspot being challenged, an entire hex is challenged, and everyone in that hex beacons and witnesses at the same time (while staying in compliance of local radio regulations). this would give the entire area some work to do to prove each other and also able to reward the entire area at once for the information the blockchain received. This is just a concept though and needs some thinking.

Please don’t, of course you can’t beacon all at the exact same but beaconing all in a certain timeframe would greatly disrupt LoRa traffic (even when staying within regulatory limits because those limits don’t really consider coordinated transmissions)

Any ETA for this HIP?

Tim | LongAP

Please don’t, of course you can’t beacon all at the exact same but beaconing all in a certain timeframe would greatly disrupt LoRa traffic (even when staying within regulatory limits because those limits don’t really consider coordinated transmissions)

We don’t have a design for it yet but have some ideas that won’t be disruptive.

hashc0de

Couldn’t have said it better than these two. #teamnosyncneeded

Without going into the details, a simple yes or no. Can you make it work? (hotspots functioning properly without being/needing to sync to the block)

Ay0hCrypto(ALCRYMIST)

Without going into the details, a simple yes or no. Can you make it work? (hotspots functioning properly without being/needing to sync to the block)

It already works on testnet, so the answer is yes

1

rawrmaan

It already works on testnet, so the answer is yes

AMAZING! TY.

haha yes. hotspots on testnet are not needing to sync. we'll need to move the miner codebase to take advantage of this and eventually drop the erlang miner (that does all the syncing/blockchain bits) from full hotspots. we're working with manufacturers to test this on testnet first as well.

so right now there are some hot spots out there that companies are calling light hot spots that do data only. once this hip is implemented would those hot spots be eligible to mine at that point or would they still just be data only hot spots?

13:11

and example would be the dragino lps8

It looks like you're asking about the Dragino LPS8 which is a Data-Only Hotspot. You will not earn HNT for PoC and the amount of HNT earned for transferring data is small. For every DC used to transfer a packet, the Data-Only Hotspot will earn 1 DCs worth of HNT. 1 DC = $0.00001. Setting up a Dragino LPS8 is a great way to build network coverage, but will not earn HNT for Proof-of-Coverage For more information, please visit: https://docs.helium.com/mine-hnt/data-only-hotspots

Data Only Hotspots | Helium Documentation

Data Only Hotspots are eligible for Network Data Transfer Rewards but not

Blackdawn101

so right now there are some hot spots out there that companies are calling light hot spots that do data only. once this hip is implemented would those hot spots be eligible to mine at that point or would they still just be data only hot spots?

Still data only

13:20

"light gateways" are sold via hip19, nothing labeled light right now will ever earn from PoC

13:20

They're classed as data only

BFGNeil - Trackpac.io

Still data only

ok thank you

so i just read this hip. Am i understanding it correctly that rewards that were being given to hotspots would move to validators with this? Considering rewards for hotspots continue to shrink is it a good idea to further punish hotspot owners with this? Or is there some mechanism to offset the hotspot rewards by increasing other operational rewards?

Tim

so i just read this hip. Am i understanding it correctly that rewards that were being given to hotspots would move to validators with this? Considering rewards for hotspots continue to shrink is it a good idea to further punish hotspot owners with this? Or is there some mechanism to offset the hotspot rewards by increasing other operational rewards?

only for creating challenges, beaconing/witness rewards stay the same

09:26

PoC Challenger 0.90% per epoch

09:26

so its not a massive change

Is this vote supposed to be a temperature check vote? I'm surprised this is already up for a full vote. The HIP PR was merged only three days ago and there are unanswered questions from the AMA.

3

BFGNeil - Trackpac.io

PoC Challenger 0.90% per epoch

it seems more and more rewards are being moved to validators while at same time more hotspots come online bringing down the full distribution of rewards for hotspots. Granted that .90% may be small on its own but combined with ever decreasing rewards this is just helping to shrink rewards further. While the suggestion does make sense from a technical point, i think we are hitting a point that the reward mechanism as a whole should be looked at and rebalanced.

2

1

wait so if I understood this right after this hip the full miners currently out don't need to be 100% synced to the blockchain everytime. Right?

1

BusyPanda

wait so if I understood this right after this hip the full miners currently out don't need to be 100% synced to the blockchain everytime. Right?

As far as I understand it, there will be no miner syncing. The P2P network will not extend to miners.

1

This POC seems to be very damaging to the intensive of deploying miners to low covered areas. Currently they can earn partly by creating challenges. By taking this away, don't you hurt the mission of global coverage?

1

i don't really want to vote yes to the hip when the validators dont stay updated and not relayed half the time, but at least i'd quit having a peer book full of denied hotspots

Tim

it seems more and more rewards are being moved to validators while at same time more hotspots come online bringing down the full distribution of rewards for hotspots. Granted that .90% may be small on its own but combined with ever decreasing rewards this is just helping to shrink rewards further. While the suggestion does make sense from a technical point, i think we are hitting a point that the reward mechanism as a whole should be looked at and rebalanced.

That's what #hip-52-lorawan-subdao will help with

09:56

Reward units in wnt can be 1000x larger to help the number bias

This HIP is going to centralize PoC challenges on validators, hence they can enforce staking of HNT if you want to be challenged

Validators are for rich folk, and now more rewards is pushed to them, taking away intensive of deploying miners

Definitely curious to that effect myself

10:01

It seems the changes are pushing people towards saturation rather than towards global spread out. If challenges move to validators, then you lose that little bit of incentive you get for being further away. Granted, the higher you go the more opportunity you have, but for those that cant get high enough to reach populated cities, would greatly lose out

once the validators controll challenges what happens to the hotspots on deny-list, do the legit hotspots still get punished when they challenge or get challenged by banned hotspots still on the chain ? as we do under the current system?

Also this vote is rather pointless, seeing validators hold the mass of HNT, and will obviously vote in favor of earning more

1

Ih8mypp

Also this vote is rather pointless, seeing validators hold the mass of HNT, and will obviously vote in favor of earning more

less validator wallets than hotspot wallets

Look at past votes and see how many votes are compared to involved HNT

true (edited)

BFGNeil - Trackpac.io

That's what #hip-52-lorawan-subdao will help with

thanks for that info. I was not aware this was being addressed.. Too much news constantly coming out to keep up haahaha

1

I do a weekly news round up on YouTube for that exact reason

We should write and propose a HIP that counts votes based on how many miners a wallet has, rather than the amount of stored HNT

2

1

I think it's important to remember that the ultimate objective of light hotspots is to scale the network and improve reliability. That is good for everyone who holds HNT, and should more than offset the 0.9% loss for PoC creation.

Busy episode tomorrow, lots of great news

AlexC

I think it's important to remember that the ultimate objective of light hotspots is to scale the network and improve reliability. That is good for everyone who holds HNT, and should more than offset the 0.9% loss for PoC creation.

i think the issue is most of us dont trust the validators to hold up to their part of this

Why not?

AlexC

Why not?

they dont stay upto date as is, the consensus groups get filled with out of date validators

10:06

some even relayed, which hurts hotspot earnings

I don't agree with that at all. The vast majority of validators are on the latest version and only a handful are relayed

BFGNeil - Trackpac.io

I do a weekly news round up on YouTube for that exact reason

just subscribed

2

AlexC

I don't agree with that at all. The vast majority of validators are on the latest version and only a handful are relayed

but that handfull ends up in our consensus groups and peerbooks daily

If anything, the rewards for challenges should remain on the hotspots IMO ... why can't we shift the challenge reward to the challengee so that all sent beacons are rewarded and you get more rewards for having witnesses of that same beacon

1

Wrath - demo.hntmonitor.com

If anything, the rewards for challenges should remain on the hotspots IMO ... why can't we shift the challenge reward to the challengee so that all sent beacons are rewarded and you get more rewards for having witnesses of that same beacon

It puts extra load, and data transfer bills onto validators running construction

10:13

In an already tight margin of earnings , it makes a lot of sense for rewarding them to do this extra workload

Just from a noob point of view, this seems like a huge centralization move. Is there somebody who can explain to be the benefits on the vision side of things (I have already an idea of the benefits of Light hotspots and the improvements of not storing the blockchain locally).

Remember lgw check in to validators with grpc so the bw usage will increase, so extra cost

It might actually decrease bandwidth. Removing the need for hotspots to sync blocks will be a huge savings.

10:15

The gRPC stuff should hopefully be pretty lightweight - but i dont know if anyone knows for sure the net difference

I just wish we could solve the saturation issue at the same time... right now this is just more incentive to squeeze into a hex to make sure you can get beacon rewards

Wrath - demo.hntmonitor.com

I just wish we could solve the saturation issue at the same time... right now this is just more incentive to squeeze into a hex to make sure you can get beacon rewards

definitely. that's fair game for a separate HIP. the existing HIP-17 variables could be tweaked to drive some of what you are looking for

1

2

Technoking

Just from a noob point of view, this seems like a huge centralization move. Is there somebody who can explain to be the benefits on the vision side of things (I have already an idea of the benefits of Light hotspots and the improvements of not storing the blockchain locally).

Unfortunately, in order to effectively construct a PoC challenge, you need a copy of the chain. Continuing to have hotspots sync the chain is not sustainable for the reasons outlined in the HIP. So, that leaves validators as the logic place for this responsibility. There are 3400+ validators owned by over 600 accounts by last check. So, I'm not sure what you mean by centralization. Becoming a validator is permissionless.

PaulM

Unfortunately, in order to effectively construct a PoC challenge, you need a copy of the chain. Continuing to have hotspots sync the chain is not sustainable for the reasons outlined in the HIP. So, that leaves validators as the logic place for this responsibility. There are 3400+ validators owned by over 600 accounts by last check. So, I'm not sure what you mean by centralization. Becoming a validator is permissionless.

so after this the current full miners that are out don't need to keep a sync from the blockchain ?

BusyPanda

so after this the current full miners that are out don't need to keep a sync from the blockchain ?

Correct. No more syncing once this is fully implemented. The extra up time for hotspot owners should be very welcome.

1

PaulM

Unfortunately, in order to effectively construct a PoC challenge, you need a copy of the chain. Continuing to have hotspots sync the chain is not sustainable for the reasons outlined in the HIP. So, that leaves validators as the logic place for this responsibility. There are 3400+ validators owned by over 600 accounts by last check. So, I'm not sure what you mean by centralization. Becoming a validator is permissionless.

To be fair, a few of those 600 accounts own a lot of them, so some centralisation concern is valid. This is not specifically related to or due to this HIP however. (edited)

2

one thing I would say that the miner not requiring to keep a blockchain copy and synced is very valuable to scale the network more specifically in developing countries. Currently in my country internet itself is not that accessible and even if I want to use a sim router to place it at a location the network is not that super great with 5g speeds and data required to keep the hotspot in sync is high. Moreover even regular broadband or fiber internet is not that much available yet and some still have data caps. I would request for people to see this hip with this aspect too. If it gives up these challengers reward but provides actual option for scalability for the network especially for places outside of developed nations. (edited)

2

BusyPanda

one thing I would say that the miner not requiring to keep a blockchain copy and synced is very valuable to scale the network more specifically in developing countries. Currently in my country internet itself is not that accessible and even if I want to use a sim router to place it at a location the network is not that super great with 5g speeds and data required to keep the hotspot in sync is high. Moreover even regular broadband or fiber internet is not that much available yet and some still have data caps. I would request for people to see this hip with this aspect too. If it gives up these challengers reward but provides actual option for scalability for the network especially for places outside of developed nations. (edited)

Thanks for pointing that out.

Will this change become inevitable at some point in the future as the network grows, regardless of whether it's adopted now?

GreatScottyMac

Will this change become inevitable at some point in the future as the network grows, regardless of whether it's adopted now?

IMO the network doesn't survive without moving challenge creation validators, so yes

10:46

We're already at the point where it needs to happen. Can't wait any longer

1

I suppose the only other option would be to somehow upgrade the capabilities of current hotspot hardware?

10:48

Which wouldn't make sense

It's not only the hardware but also the networking, as illustrated by @BusyPanda above

2

Yeah exactly, it would increase the hotspot cost and make all the current ones obsolete. Even then, you'd have the p2p network issues to deal with, which would get worse over time with more hotspots no matter how good the hardware is

It's really quite remarkable that we have a 500k+ node p2p network actually working.

3

2

1

So the NO votes are probably coming from people who only want to be in this for the short term

‍♂️

rawrmaan

IMO the network doesn't survive without moving challenge creation validators, so yes

I disagree, the challenges are not the issue, the networking is. The correct statement (imo) would be that the network doesn't survive without fixing the networking problems. One of the solutions is light hotspots but to do that you need challenging at the validators.

hashc0de

It's really quite remarkable that we have a 500k+ node p2p network actually working.

It has to be the largest P2P network in the world, surely?

groot

I disagree, the challenges are not the issue, the networking is. The correct statement (imo) would be that the network doesn't survive without fixing the networking problems. One of the solutions is light hotspots but to do that you need challenging at the validators.

Technically correct, the hotspots need to no longer connect to the p2p network, but they technically still could create challenges. The degree of complexity there is unknown to me

GreatScottyMac

So the NO votes are probably coming from people who only want to be in this for the short term

‍♂️

As a devil's advocate, there are multiple ways to go about moving the challengers to validators. My concerns at least are with the design not the fact of moving. I think there should have been more discussion on the specifics. This is being sold as a vote on light hotspots vs. not. It's much more than that.

1

AlexC

It has to be the largest P2P network in the world, surely?

pretty close. bittorrent is maybe bigger.

hashc0de

pretty close. bittorrent is maybe bigger.

But is that one network?

1

Yeah i guess the difference with bittorrent is that you don't have a shared state for all nodes

1

I mean, rewards from creating challenges are such a tiny fraction of total earnings, I would think that avoiding even one episode of falling out of sync would make it worthwhile to hand challenge creation over to validators

True, but challenge creation is the only reward for hotspots that are lone wolves. It's a UBI of sorts. Psychologically it may keep many hotspots running before they have witnesses (edited)

10:55

We can't know the true effect of losing this UBI until it's gone, unfortunately. It's not measurable

rawrmaan

True, but challenge creation is the only reward for hotspots that are lone wolves. It's a UBI of sorts. Psychologically it may keep many hotspots running before they have witnesses (edited)

Ahh, yes, I didn't think about that. This is where the no votes are coming from

The wording of the vote announcement seems to come down to: "vote for this or no light hotspots for you". It isn't very neutral and I think that undermines the 'HIP it all' attitude that Helium is taking. It is simply not true that you have to agree with this HIP, in this form, in order to get light hotspots so it shouldn't be brought that way.

1

PaulM

As a devil's advocate, there are multiple ways to go about moving the challengers to validators. My concerns at least are with the design not the fact of moving. I think there should have been more discussion on the specifics. This is being sold as a vote on light hotspots vs. not. It's much more than that.

Exactly...For example: We move challenging to validators, what happens when some1 joins the network? Are they force to delegate in order to get challenges?

1

Drunfos

Exactly...For example: We move challenging to validators, what happens when some1 joins the network? Are they force to delegate in order to get challenges?

Certainly not, since delegation only exists in 10,000 HNT chunks

groot

The wording of the vote announcement seems to come down to: "vote for this or no light hotspots for you". It isn't very neutral and I think that undermines the 'HIP it all' attitude that Helium is taking. It is simply not true that you have to agree with this HIP, in this form, in order to get light hotspots so it shouldn't be brought that way.

Well said.

How certain are you? They didnt specify anything. Maybe they lower staking entry and "force" everyone to stake in order to witness. As staking services are paid, you pay for the hotspot+service and we move to private network..No more peoples network..:p (edited)

groot

The wording of the vote announcement seems to come down to: "vote for this or no light hotspots for you". It isn't very neutral and I think that undermines the 'HIP it all' attitude that Helium is taking. It is simply not true that you have to agree with this HIP, in this form, in order to get light hotspots so it shouldn't be brought that way.

It has also been rather hasty. Concerns have been raised but not properly addressed. I was surprised to see the vote happening so soon.

groot

The wording of the vote announcement seems to come down to: "vote for this or no light hotspots for you". It isn't very neutral and I think that undermines the 'HIP it all' attitude that Helium is taking. It is simply not true that you have to agree with this HIP, in this form, in order to get light hotspots so it shouldn't be brought that way.

any suggestions?

BFGNeil - Trackpac.io

any suggestions?

For starters I would remove the erroneous causality between HIP55 and light hotspots. We're not voting on light hotspots, we are voting for a proposal that is one of the paths that will lead to light hotspots. Let me be clear, I am not arguing the content of the HIP, but rather the way that it is presented.

1

sure, just to be clear i believe some are missreading it as all poc is moving

11:10

and suggested maybe an edit, so if you have any feedback as to what, I'll do a pr

i dont think it is all poc, as much as how much power/control over the network validators end up with in the end. and how reliable the network becomes with validators holding the PoC creation power.

groot

For starters I would remove the erroneous causality between HIP55 and light hotspots. We're not voting on light hotspots, we are voting for a proposal that is one of the paths that will lead to light hotspots. Let me be clear, I am not arguing the content of the HIP, but rather the way that it is presented.

Not sure how moving Challenges to Validators is not required for light hotspots?

just label it as " solution to hotspot sync issues and sd card storage" and everyone will vote yes

(edited)

2

1

BusyPanda

just label it as " solution to hotspot sync issues and sd card storage" and everyone will vote yes

(edited)

AKA "No more Relay or Port forwarding" HIP.

lol

ElonTusk | 5Gchipped

AKA "No more Relay or Port forwarding" HIP.

lol

yeah. that too.

11:30

like seriously in a way this hip actually solves so much for the people with gateway I just don't understand why someone with gateway will vote against it

1

ElonTusk | 5Gchipped

Not sure how moving Challenges to Validators is not required for light hotspots?

You seem to be missing the point, this HIP, in this form, is not the only road to light hotspots. Even if we accept that we have to move the challenges to validators this is not the only way to do it and light hotspots are not dependent on this particular implementation. I am not arguing the other methods are preferable or even viable, but it shouldn't be presented as a 'this or nothing' case. (edited)

PSA: https://heliumvote.com/ Go vote, and make your voice heard.

Helium Vote

Helium Vote is where the Helium Community comes together to make decisions on the Network.

1

groot

You seem to be missing the point, this HIP, in this form, is not the only road to light hotspots. Even if we accept that we have to move the challenges to validators this is not the only way to do it and light hotspots are not dependent on this particular implementation. I am not arguing the other methods are preferable or even viable, but it shouldn't be presented as a 'this or nothing' case. (edited)

Sorry, was in a work meeting.

What is the other road then? How are LHS possible without moving Challenges away from them (and all the related mechanisms that go with that)?

ElonTusk | 5Gchipped

Sorry, was in a work meeting.

What is the other road then? How are LHS possible without moving Challenges away from them (and all the related mechanisms that go with that)?

There are other options with regard to the specifics. For example, whether or not to move the 0.9% to the validators, and also whether specifically the consensus group validators should be the ones creating challenges or spreading the challenge creation around to all validators. I noted a concern with having only 43 at a time of the 3400 validators taking this one.

PaulM

There are other options with regard to the specifics. For example, whether or not to move the 0.9% to the validators, and also whether specifically the consensus group validators should be the ones creating challenges or spreading the challenge creation around to all validators. I noted a concern with having only 43 at a time of the 3400 validators taking this one.

IMO the 0.9% is only fair. They are doing the work, they should get that reward.

PaulM

There are other options with regard to the specifics. For example, whether or not to move the 0.9% to the validators, and also whether specifically the consensus group validators should be the ones creating challenges or spreading the challenge creation around to all validators. I noted a concern with having only 43 at a time of the 3400 validators taking this one.

I noted a concern with having only 43 at a time of the 3400 validators taking this one.

Sorry, been a lot going on. Could you remind us what that concern was?

In order to do PoC generation, the hotspots have to sync the chain, right? As a hotspot owner, I'd really prefer the hotspots to not have to stay in sync, I "give up" more than the equivalent of PoC rewards in witnessing and beaconing rewards by having hotspots that struggle to stay in sync.

2

ElonTusk | 5Gchipped

I noted a concern with having only 43 at a time of the 3400 validators taking this one.

Sorry, been a lot going on. Could you remind us what that concern was?

CG members already (before the PoC interval reduction) struggle to keep up with CG work, even those on expensibve highly spec'd dedicated servers. The HIP-55 design adds work to CG members and has the potential for a ton of incoming connections from a hex to check if they're in CG. The concern is reducing block production performance, I think. (edited)

ElonTusk | 5Gchipped

I noted a concern with having only 43 at a time of the 3400 validators taking this one.

Sorry, been a lot going on. Could you remind us what that concern was?

I commented here: https://github.com/helium/HIP/issues/362

1

ElonTusk | 5Gchipped

I noted a concern with having only 43 at a time of the 3400 validators taking this one.

Sorry, been a lot going on. Could you remind us what that concern was?

Also, all the light hotspots are going to dependent on validators for services - even those that are not actively creating challenges. However, there is no additional incentive for those validators. What if all the validator operators choose not to offer that since it is extra cost to them without any compensation? Then you are going to have 500K hotspots dependent on a few validators for those services. That could be just as bad as the p2p network depending on how generous the validators are feeling.

3

PaulM

Also, all the light hotspots are going to dependent on validators for services - even those that are not actively creating challenges. However, there is no additional incentive for those validators. What if all the validator operators choose not to offer that since it is extra cost to them without any compensation? Then you are going to have 500K hotspots dependent on a few validators for those services. That could be just as bad as the p2p network depending on how generous the validators are feeling.

What if all the validator operators choose not to offer that since it is extra cost to them without any compensation? If the light hotspots do not get challenged anymore, the whole blockchain becomes useless as well as the HNT coin. So it would impact the validators badly too

agourk

What if all the validator operators choose not to offer that since it is extra cost to them without any compensation? If the light hotspots do not get challenged anymore, the whole blockchain becomes useless as well as the HNT coin. So it would impact the validators badly too

Long-term, yes. Unfortunately, as with any system, there are still many people that take a short-term, selfish view.

let me understand this right -> Today, the Hotspots creating PoC Challenges and submitting receipts to the blockchain are rewarded with 0.9% of HNT rewarded per epoch. We propose that this subsidy be moved to the Validator Challenger

12:29

means we will produce even less

1

12:29

more for the validators...

It’s such a cheap tradeoff for no syncing honestly

12:31

Sacrificing 0.01 per day/hotspot?

12:32

Let them be compensated for lifting the weights (edited)

4

1

depends on the numbers, i voted against, lately all work made to "ease" the network is done to the benefit of partners etc

Dejavu

let me understand this right -> Today, the Hotspots creating PoC Challenges and submitting receipts to the blockchain are rewarded with 0.9% of HNT rewarded per epoch. We propose that this subsidy be moved to the Validator Challenger

my hotspot has gone offline often because of syncing issues, and I think it cost me more than 0.9% of my rewards

1

agourk

my hotspot has gone offline often because of syncing issues, and I think it cost me more than 0.9% of my rewards

not my fault that your manufacturer shipped you the cheapest hardware

1

12:35

and dont know/want to fix the fw (edited)

Tarek

Let them be compensated for lifting the weights (edited)

https://discord.com/channels/404106811252408320/942804859768438834/943962303450189874 PaulM said validators dont get compensation

then the hip must be edited

12:37

validators are independent, i see bad things comming from this

12:37

see what happened when the block got stuck

12:37

begging people to run commands

12:38

i dont trust validators, sorry

2

Dejavu

and dont know/want to fix the fw (edited)

funny, so, what if those having the most issues get their FW straight from the helium team?

timamook_psap35

funny, so, what if those having the most issues get their FW straight from the helium team?

actually very few, og, rak and sensecap

Dejavu

i dont trust validators, sorry

Both ways validators control the network

12:39

They could stop now and the network would shutdown

12:39

No more blocks

Tarek

Both ways validators control the network

exactly

12:39

give them even more control (edited)

12:39

lets remember the last "400" added validators

12:40

was some discussion there

agourk

https://discord.com/channels/404106811252408320/942804859768438834/943962303450189874 PaulM said validators dont get compensation

There are two new roles of validators described in the hip - one for creating challenges. that is compensated. and one for providing other services like notifying hotspots they've been challenged, providing routing information for data packets, etc. the latter has no incentive mechanism in the proposal

Dejavu

give them even more control (edited)

They’re already in control enough to shut down the network

12:41

I don’t see why letting them handle the challenges is a big deal

Tarek

They’re already in control enough to shut down the network

im just saying that giving to much power to independent people is not good

1

Dejavu

im just saying that giving to much power to independent people is not good

But...It's the people's network! It's decentralized!

Validators are people too! (edited)

1

Dejavu

im just saying that giving to much power to independent people is not good

Well it’s the peoples network

yeaa.. good one :))

If the people decide to shut it down then oh well

yea, someone make a hip

Dejavu

yea, someone make a hip

You're a people too. You can write a hip.

1

Peso die Schildkröte 🇩🇪 pinned a message to this channel. 02/17/2022 12:46 PM

krby

You're a people too. You can write a hip.

too busy fixing cotx hotspots

1

What if this was moved to a whole new "miner" so to say... one that any community member could run that stays in sync with the chain and just issues challenges?

13:00

Get rewards for the challenges submitted and up time!?

13:01

That way, poc remains completely decentralized and not driven by only those with the pockets to run a validator

there are multiple solution options, but they have to be balanced with the end goal, difficulty in implementation etc.

Wrath - demo.hntmonitor.com

That way, poc remains completely decentralized and not driven by only those with the pockets to run a validator

IMO the requirements for this "PoC creation miner" would have to be: - Reliable network connection that allows inbound connections (port-forwarding) - Enough CPU and disk (speed and capacity) to stay in sync. What hardware would support that? It feels a lot like a VPS instance running on AWS or similar. Would you pay $40/mo to run that? Sure it COULD be run at home, but then without port forwarding you're back to p2p network issues. You also wouldn't want it to be an RPi with an SD card, then you're stuck managing space and wear level like the current hotspots are, not counting the CPU load. (edited)

Sounds like something I'd just run on my validator servers (edited)

1

krby

IMO the requirements for this "PoC creation miner" would have to be: - Reliable network connection that allows inbound connections (port-forwarding) - Enough CPU and disk (speed and capacity) to stay in sync. What hardware would support that? It feels a lot like a VPS instance running on AWS or similar. Would you pay $40/mo to run that? Sure it COULD be run at home, but then without port forwarding you're back to p2p network issues. You also wouldn't want it to be an RPi with an SD card, then you're stuck managing space and wear level like the current hotspots are, not counting the CPU load. (edited)

i think that is a more viable option for the subdao chains, and we should consider layer two nodes, but that's a totally different set of hips, and a new hip

AlexC

Sounds like something I'd just run on my validator servers (edited)

okay what's your opinion running a validator about this HIP ? for or against and why?

krby

IMO the requirements for this "PoC creation miner" would have to be: - Reliable network connection that allows inbound connections (port-forwarding) - Enough CPU and disk (speed and capacity) to stay in sync. What hardware would support that? It feels a lot like a VPS instance running on AWS or similar. Would you pay $40/mo to run that? Sure it COULD be run at home, but then without port forwarding you're back to p2p network issues. You also wouldn't want it to be an RPi with an SD card, then you're stuck managing space and wear level like the current hotspots are, not counting the CPU load. (edited)

I think the trade off is the same... what you just described to me, is MUCH easier for me to solve than 10,000 HNT

krby

IMO the requirements for this "PoC creation miner" would have to be: - Reliable network connection that allows inbound connections (port-forwarding) - Enough CPU and disk (speed and capacity) to stay in sync. What hardware would support that? It feels a lot like a VPS instance running on AWS or similar. Would you pay $40/mo to run that? Sure it COULD be run at home, but then without port forwarding you're back to p2p network issues. You also wouldn't want it to be an RPi with an SD card, then you're stuck managing space and wear level like the current hotspots are, not counting the CPU load. (edited)

Also, it wouldnt been that difficult to do honestly, for any normal operator... I mean we already have a ton of people that know how to actually port forward! The hardware requirements cant be THAT daunting considering that your tiny little hotspot is doing it right now!!

1

Wrath - demo.hntmonitor.com

I think the trade off is the same... what you just described to me, is MUCH easier for me to solve than 10,000 HNT

Sure, but is it financially worth it to do it? $40/mo (or $20 for a real small instance) for how many rewards?

How do you make them sybil resistant? You need some kind of stake

krby

Sure, but is it financially worth it to do it? $40/mo (or $20 for a real small instance) for how many rewards?

It is for me, cause i have servers already that i can spin up that can handle this noproblem...plus the bandwidth

Wrath - demo.hntmonitor.com

Also, it wouldnt been that difficult to do honestly, for any normal operator... I mean we already have a ton of people that know how to actually port forward! The hardware requirements cant be THAT daunting considering that your tiny little hotspot is doing it right now!!

They're doing it, poorly. Many hotspots don't stay in sync, or aren't reachable because of inbound connections.

krby

They're doing it, poorly. Many hotspots don't stay in sync, or aren't reachable because of inbound connections.

Right but they, again, are tiny in comparison to even a laptop with an icore 5

13:25

which is shit lol

13:26

i bet an icore 3 on an ssd could outperform the hotspot in challenges

Wrath - demo.hntmonitor.com

It is for me, cause i have servers already that i can spin up that can handle this noproblem...plus the bandwidth

Ahh, yes. I see. So...for the group that can do this, sure! It's both the network connectivity (and home ISP / router combos suck for this) and the compute / disk.

Wrath - demo.hntmonitor.com

i bet an icore 3 on an ssd could outperform the hotspot in challenges

Totally agree, tiny "real" computer with not an SD card can keep up with the chain just fine.

It would just be a really nice option, to not have to be rich in order to run this part of the code is my real feedback

13:27

plus it gives the opportunity for both the rich or savvy to participate

I tried a raspberry pi with iscsi storage once, it could also keep up with the chain without any issues, no absorb spikes, just decent performance. The SD card is the real buzzkill.

1

that's a nice incentive, imo... a PI would totally make your money's worth for keeping it online

Wrath - demo.hntmonitor.com

plus it gives the opportunity for both the rich or savvy to participate

The problem to solve (as @AlexC ) pointed out is that we'd need enough people doing this, or some "stake" in the game so the PoC generation isn't game-able.

1

krby

The problem to solve (as @AlexC ) pointed out is that we'd need enough people doing this, or some "stake" in the game so the PoC generation isn't game-able.

I agree 100% ... I dont think the answer im giving is to deny validators... it's to allow more than validators to be able to do this simple part of the poc

13:31

So if you want to run it on a val... go ffor it, but it shouldnt be a requirement to be a val just to poc

13:31

errr, issue a challenge i mean

Wrath - demo.hntmonitor.com

I agree 100% ... I dont think the answer im giving is to deny validators... it's to allow more than validators to be able to do this simple part of the poc

Ahhh, ok. So if a val operator wants to be rewarded for PoC challenge work, they can opt in, but anyone else who thinks they can make this work can also, without a high (or any?) stake required.

1

Exactly that, just have the know how, the equipment, and participate.... what would be even more awesome, is eventually with a model like this, validators could essentially utilize in field poc challengers or themselves to further expand their compute and knock out blocks at a time

So let me spin up 100k instances and get all the 0.9%.

A PoC-specific role with some low stake requirement would probably work (edited)

13:35

100 HNT, say

groot

So let me spin up 100k instances and get all the 0.9%.

i would hope not lol that would be wasteful for you i would think

13:36

no possible way .9% of rwards could pay for 100K of an overload of servers

Wrath - demo.hntmonitor.com

i would hope not lol that would be wasteful for you i would think

Not if "get all the 0.9%" really means "can game PoC generation for my cheating miner farm"

1

Krby is getting the subtext

krby

Not if "get all the 0.9%" really means "can game PoC generation for my cheating miner farm"

So, how is it done now... its isnt as if owning more hotspots makes you challenge more right? (edited)

13:37

it's essentially the same no?

13:37

if i bought 100K hotspots do I get all the 0.9% ?

So...ya I guess that means the stake (when functioning as some friction for acquiring too many) for PoC is naturally the cost of a hotspot?

essentially, but im pretty sure even hotspots have a limitation to when they can issue a challenge.. i would imagine the new service would be no different

In my example the 0.9% wasn’t really the main goal, the main goal was being in charge of challenge creation, just to clarify.

1

that is the same as owning 100K hotspots though bro

13:39

you're just moving the service around

13:39

doesnt matter if its a hotspot or servers... you can do that now if that was the argument

13:40

well 100K hotspots would be more profitable since you can actually get other rewards

13:40

I dont see someone gaming for poc rewards though

Now that we added staking to the mix it is the same, yes. My example was quickly destroyed due to the staking

1

Anyway, good discussion, love it guys.. lots a good topics on this one

13:41

I gotta head out for a bit but ill check in every once in a while

1

Dejavu

let me understand this right -> Today, the Hotspots creating PoC Challenges and submitting receipts to the blockchain are rewarded with 0.9% of HNT rewarded per epoch. We propose that this subsidy be moved to the Validator Challenger

Become a validator or stake haha

Mining Chef

Become a validator or stake haha

just do both

So tell me how to vote .. yes or no

Mining Chef

So tell me how to vote .. yes or no

I will not, but I voted yes.

I have 90,000 hnt so my vote matters... A little haha

1

Mining Chef

I have 90,000 hnt so my vote matters... A little haha

14:31

lol

There are many voting against this because they don't want to lose the challenger rewards... I don't think they have any idea how many witnesses and beacons are being lost simply because of the p2p issues!!! Assuming the validators are up to the task, a properly operating network should more than make up for the challenger losses.

2

Hashcode and the Professor know more then me... So I think I trust them haha

1

malder

There are many voting against this because they don't want to lose the challenger rewards... I don't think they have any idea how many witnesses and beacons are being lost simply because of the p2p issues!!! Assuming the validators are up to the task, a properly operating network should more than make up for the challenger losses.

Losing 0.9% is nothing. Very cheap price to pay for fixing so many issues.

1

I'd be a fan of imposing some kind of minimum service agreement on validators though. I know that's philosophically difficult... but hotspot owners can't build their own hardware, and that's fair... we want a predictably functional network. Likewise if the network is to be run by validators, they should meet some minimum requirements for hardware, network access, and uptime.

1

malder

I'd be a fan of imposing some kind of minimum service agreement on validators though. I know that's philosophically difficult... but hotspot owners can't build their own hardware, and that's fair... we want a predictably functional network. Likewise if the network is to be run by validators, they should meet some minimum requirements for hardware, network access, and uptime.

Just like SC Prime

malder

I'd be a fan of imposing some kind of minimum service agreement on validators though. I know that's philosophically difficult... but hotspot owners can't build their own hardware, and that's fair... we want a predictably functional network. Likewise if the network is to be run by validators, they should meet some minimum requirements for hardware, network access, and uptime.

Well to be fair there is already a punishment system in place for underperforming validators.

2

14:35

Might need some adjustments though.

malder

There are many voting against this because they don't want to lose the challenger rewards... I don't think they have any idea how many witnesses and beacons are being lost simply because of the p2p issues!!! Assuming the validators are up to the task, a properly operating network should more than make up for the challenger losses.

i hope they are since their vote's easily out weigh the very few hotspot owners who actually vote on hips

ElonTusk | 5Gchipped

Well to be fair there is already a punishment system in place for underperforming validators.

that system sucks

Its not about the rewards..Its about centralizing challenge generation

1

Drunfos

Its not about the rewards..Its about centralizing challenge generation

by the vote of the validators they vote to assume the work, and the power, so the rest of the "people" in the people's network are now at their mercy

Drunfos

Its not about the rewards..Its about centralizing challenge generation

"Centralized" over thousands of validators? lol

Ay0hCrypto(ALCRYMIST)

that system sucks

Can you expand on that?

ElonTusk | 5Gchipped

"Centralized" over thousands of validators? lol

a thousand vs a half million yes that is centralized

Ay0hCrypto(ALCRYMIST)

a thousand vs a half million yes that is centralized

No it's not.

ElonTusk | 5Gchipped

Can you expand on that?

there is no positive action in that system just negatives

ElonTusk | 5Gchipped

No it's not.

if ya'll are upto the task, then it works for me. the people have the option to form together and fund their own validators

1

Ay0hCrypto(ALCRYMIST)

there is no positive action in that system just negatives

Yes there is. Positive: You're in CG and make rewards.

Ay0hCrypto(ALCRYMIST)

if ya'll are upto the task, then it works for me. the people have the option to form together and fund their own validators

Ya'll? I'm not a validator

By acknowledging that you need a validator to emit challenges, you are complying to a model where validators can exert other types of pressures on the rest of the network

ElonTusk | 5Gchipped

Yes there is. Positive: You're in CG and make rewards.

that's a positive reward, not in the grading of that penalty score.

ElonTusk | 5Gchipped

Ya'll? I'm not a validator

they*

1

Yeah, lots of people set up validator pools

ElonTusk | 5Gchipped

Yeah, lots of people set up validator pools

we've been organizing one.

1

14:46

figure 50 of us split the stake

14:46

assign a board, and run it

Ay0hCrypto(ALCRYMIST)

figure 50 of us split the stake

Nice!

NA joint venture

14:47

only reason i dont rally against this, as it doesnt stop the people from taking what control they desire to dedicate the time and resources too

14:47

it only becomes an us(hotspots) vs them (validators) if the hotspot owners let it

14:48

nothing stops a validator from buying hotspots, and vice versa

1

And with hip55 nothing will stop a validator from putting a minimal staking fee for miners to use the network (example)

Drunfos

And with hip55 nothing will stop a validator from putting a minimal staking fee for miners to use the network (example)

Sorry, not following what you're trying to say.

If you need a challenge to beacon, and you need to beacon to witness (hip54)

Drunfos

And with hip55 nothing will stop a validator from putting a minimal staking fee for miners to use the network (example)

You know the validators could stop all POC rewards already, right?

So your saying that theres no intent as of now

Drunfos

So your saying that theres no intent as of now

The validators are not going to charge hotspots to process POC

Drunfos

And with hip55 nothing will stop a validator from putting a minimal staking fee for miners to use the network (example)

No. Hotspots will have a choice of what validator that use for typical operations.

1

PaulM

No. Hotspots will have a choice of what validator that use for typical operations.

That too

If a validator if ignoring your hotspot, you can always switch

@Drunfos the network is broken. It's not working. If you want a fully distributed model then you can have a perfectly equal share of nothing. I can't offer a technical alternative, so I'm comfortable allowing those with the resources and expertise to take a fair share for the value they provide, and I get to share in that success proportionately with what I provide (a rooftop and minimal investment).

2

PaulM

If a validator if ignoring your hotspot, you can always switch

Just to be clear, a validator doing that would trigger penalties, right?

ElonTusk | 5Gchipped

Just to be clear, a validator doing that would trigger penalties, right?

Not initially. There was talk of down the road, using "attestation" information to either further reward validators for that service or punish them for not doing it well (edited)

1

15:00

This goes back to my concern. The hip adds work for validators to support light hotspots (in addition and separate from the challenge creation) which has no incentive system in this proposal

malder

@Drunfos the network is broken. It's not working. If you want a fully distributed model then you can have a perfectly equal share of nothing. I can't offer a technical alternative, so I'm comfortable allowing those with the resources and expertise to take a fair share for the value they provide, and I get to share in that success proportionately with what I provide (a rooftop and minimal investment).

You can, and you should, try and compreend different aproaches to the proposal. Im actually here to engage and enjoying the responses @PaulM is giving..

PaulM

This goes back to my concern. The hip adds work for validators to support light hotspots (in addition and separate from the challenge creation) which has no incentive system in this proposal

How so? It's the same work hotspots are doing now. Why would there be additional rewards for the same work?

Wouldn’t this change indirectly deincentivise isolated hotspot placement as challenging rewards will be lost?

PaulM

This goes back to my concern. The hip adds work for validators to support light hotspots (in addition and separate from the challenge creation) which has no incentive system in this proposal

I wouldn't say there is "no" incentive. The incentive is to do the work with a high level of service in order to maximize the performance and uptime of the network. Since all validators are by definition large HNT holders, they should want to maximize the value of HNT by keeping the network running smoothly.

1

CFHeadphase

Wouldn’t this change indirectly deincentivise isolated hotspot placement as challenging rewards will be lost?

Yep. "Lone Wolf" installs will get nothing but DC rewards.

ElonTusk | 5Gchipped

How so? It's the same work hotspots are doing now. Why would there be additional rewards for the same work?

Light hotspots will depend on validators for packet routing information, notifying them of when their hex was challenged and which CG member to send PoC/witnesses to. They do all that themselves today.

ElonTusk | 5Gchipped

Yep. "Lone Wolf" installs will get nothing but DC rewards.

Which will likely be none… (edited)

PaulM

Light hotspots will depend on validators for packet routing information, notifying them of when their hex was challenged and which CG member to send PoC/witnesses to. They do all that themselves today.

Right... so same rewards.

rawrmaan

I wouldn't say there is "no" incentive. The incentive is to do the work with a high level of service in order to maximize the performance and uptime of the network. Since all validators are by definition large HNT holders, they should want to maximize the value of HNT by keeping the network running smoothly.

Yes, I subscribe to that philosophy. I can't speak for all validator operators. We've seen with PoC, if there is a way to game, people will do it.

CFHeadphase

Which will likely be none… (edited)

Yep. Usefulness to the network = rewards.

Wider coverage is useful to the network

15:04

Having 30 hotspots in one hex is the opposite of useful

Also, we've seen poorly configured validators. Without a direct incentive system (penalties or rewards), there is no signal for individual validator operators to configure their setup correct to support validators

CFHeadphase

Wider coverage is useful to the network

Only if that coverage is used. You don't get rewards just for existing.

CFHeadphase

Wouldn’t this change indirectly deincentivise isolated hotspot placement as challenging rewards will be lost?

Is there really any incentive now? What's the ROI on a hotspot earning only challenger rewards? Far far better in that situation to buy two hotspots and have them witness each other with maximum witness rewards. And those rewards would be more reliable (and should be higher) after this HIP.

1

PaulM

Yes, I subscribe to that philosophy. I can't speak for all validator operators. We've seen with PoC, if there is a way to game, people will do it.

I think we can pretty safely assume that most validator operators will not be going out of their way to not provide gRPC. It's just really hard to cryptographically verify this stuff. I think at some level there is an implicit trust/good will component to running a validator

2

ElonTusk | 5Gchipped

Only if that coverage is used. You don't get rewards just for existing.

People won’t adopt Helium if coverage isn’t already there

PaulM

Also, we've seen poorly configured validators. Without a direct incentive system (penalties or rewards), there is no signal for individual validator operators to configure their setup correct to support validators

Agreed. Underperforming Validators need to be penalized in favor of well performing ones.

rawrmaan

I think we can pretty safely assume that most validator operators will not be going out of their way to not provide gRPC. It's just really hard to cryptographically verify this stuff. I think at some level there is an implicit trust/good will component to running a validator

At minimum, it will require opening another port that is not required today.

malder

Is there really any incentive now? What's the ROI on a hotspot earning only challenger rewards? Far far better in that situation to buy two hotspots and have them witness each other with maximum witness rewards. And those rewards would be more reliable (and should be higher) after this HIP.

Sure, but hotspot supply is still severely limited

CFHeadphase

People won’t adopt Helium if coverage isn’t already there

So install more hotspots.

I think it's safe to assume that most validator operators will open the port. And if/when we see that they refuse to do so, we can take it from there. As Abhay said on the call, in this scenario, Helium Inc. will gladly serve the gRPC traffic until a better solution is invented

1

CFHeadphase

Sure, but hotspot supply is still severely limited

Not really

ElonTusk | 5Gchipped

So install more hotspots.

Bobcat’s lead time is now 30 weeks

And light hotspots opens the floodgates for more hotspots.

ElonTusk | 5Gchipped

And light hotspots opens the floodgates for more hotspots.

That’s true

1

CFHeadphase

Bobcat’s lead time is now 30 weeks

You can literally order sensecaps that will arrive in March right now.

‍♂️

rawrmaan

I think it's safe to assume that most validator operators will open the port. And if/when we see that they refuse to do so, we can take it from there. As Abhay said on the call, in this scenario, Helium Inc. will gladly serve the gRPC traffic until a better solution is invented

I hope you are right. I really want light hotspots. I also want to make sure validators have the proper incentives to support them. The network shouldn't have to rely on Helium Inc running infrastructure.

3

As an aside, did anyone get a notification from the app about these votes? I did not.

ElonTusk | 5Gchipped

As an aside, did anyone get a notification from the app about these votes? I did not.

Should be rolling out. I personally haven’t gotten it yet.

1

Joey

Should be rolling out. I personally haven’t gotten it yet.

I'll try to remember to raise the flag if something doesn't come in by tomorrow.

1

15:23

(added to my calendar so I'll actually remember.

)

ElonTusk | 5Gchipped

As an aside, did anyone get a notification from the app about these votes? I did not.

Not me. Gonna be even worse when each vendor has their own app

So, PoC is what originally attracted many miners to helium...this was similar to Pow in asic mining. I think if this HIP 55 is approved, the validators will have an overwhelming amount of power and responsibility. We will lose decentralization. (edited)

Sociotter

So, PoC is what originally attracted many miners to helium...this was similar to Pow in asic mining. I think if this HIP 55 is approved, the validators will have an overwhelming amount of power and responsibility. We will lose decentralization. (edited)

We just discussed this. They already do. They could shut down everything if they wished, but they can't and won't. Because it's decentralized...

I just voted against this

ElonTusk | 5Gchipped

We just discussed this. They already do. They could shut down everything if they wished, but they can't and won't. Because it's decentralized...

@ElonTusk | 5Gchipped Exactly, THEY could SHUT DOWN EVERYTHING. That is sort of the problem right there.

Sociotter

@ElonTusk | 5Gchipped Exactly, THEY could SHUT DOWN EVERYTHING. That is sort of the problem right there.

You need to read the whole post

It seems like every HIP plan screws the miners

1

Crypto_Cowboy

I just voted against this

Why, if you don't mind me asking?

1

Sociotter

@ElonTusk | 5Gchipped Exactly, THEY could SHUT DOWN EVERYTHING. That is sort of the problem right there.

That has always, and will always, be the case. That's how any blockchain works.

ElonTusk | 5Gchipped

Why, if you don't mind me asking?

No more challenge rewards

Crypto_Cowboy

It seems like every HIP plan screws the miners

How are we screwed? Light hotspots fixed a ton of problems that affects miners.

Crypto_Cowboy

No more challenge rewards

Dude... 0.9% is nothing.

ElonTusk | 5Gchipped

How are we screwed? Light hotspots fixed a ton of problems that affects miners.

Care to enlighten me? I haven’t had any trouble.

rawrmaan

That has always, and will always, be the case. That's how any blockchain works.

Exactly...Seems similar to Etherium's switch from PoW to PoS...centralization of the wealthy at the cost of the poor!

1

Crypto_Cowboy

Care to enlighten me? I haven’t had any trouble.

p2p, relays, port forwarding, the 30%-50% of poc that is currently dropped...

We really should find ways of increasing the rewards of the Hotspots which halved 3 times on top of the halvening in August. This is the opposite taking responsibility and rewards away from the hotspots. If we can’t find a way to increase the rewards it won’t be profitable by the actual rate in 2-3 Months. And I need to shit down my fleet of Hotspots and find some idiots that didn’t get the news that it’s not worth it anymore to sell em all.

1

Sociotter

Exactly...Seems similar to Etherium's switch from PoW to PoS...centralization of the wealthy at the cost of the poor!

This changes nothing about the consensus algorithm. It's nothing like the change you're citing

DaBlawb - Flere Ego Te Judico⚡

We really should find ways of increasing the rewards of the Hotspots which halved 3 times on top of the halvening in August. This is the opposite taking responsibility and rewards away from the hotspots. If we can’t find a way to increase the rewards it won’t be profitable by the actual rate in 2-3 Months. And I need to shit down my fleet of Hotspots and find some idiots that didn’t get the news that it’s not worth it anymore to sell em all.

This is what I’m worried about.

1

DaBlawb - Flere Ego Te Judico⚡

We really should find ways of increasing the rewards of the Hotspots which halved 3 times on top of the halvening in August. This is the opposite taking responsibility and rewards away from the hotspots. If we can’t find a way to increase the rewards it won’t be profitable by the actual rate in 2-3 Months. And I need to shit down my fleet of Hotspots and find some idiots that didn’t get the news that it’s not worth it anymore to sell em all.

The way to increase the reward is by getting more devices on the network that will use data credits and thus cause increased compensation by raising HNT's value. For anyone who has a miner, your rewards in HNT terms will always decrease over time, from now until infinity. That's how the economics of this network have always been (edited)

4

DaBlawb - Flere Ego Te Judico⚡

We really should find ways of increasing the rewards of the Hotspots which halved 3 times on top of the halvening in August. This is the opposite taking responsibility and rewards away from the hotspots. If we can’t find a way to increase the rewards it won’t be profitable by the actual rate in 2-3 Months. And I need to shit down my fleet of Hotspots and find some idiots that didn’t get the news that it’s not worth it anymore to sell em all.

The only saving grace is that there are a lot of people who can’t/won’t set their miners up properly and those who can will continue to reap most of the rewards.

DaBlawb - Flere Ego Te Judico⚡

We really should find ways of increasing the rewards of the Hotspots which halved 3 times on top of the halvening in August. This is the opposite taking responsibility and rewards away from the hotspots. If we can’t find a way to increase the rewards it won’t be profitable by the actual rate in 2-3 Months. And I need to shit down my fleet of Hotspots and find some idiots that didn’t get the news that it’s not worth it anymore to sell em all.

Yeah, increase the price of HNT. Increasing the rewards will just devalue the coin.

rawrmaan

The way to increase the reward is by getting more devices on the network that will use data credits and thus cause increased compensation by raising HNT's value. For anyone who has a miner, your rewards in HNT terms will always decrease over time, from now until infinity. That's how the economics of this network have always been (edited)

And this happens way to slow, that’s why we need chainvars to increase PoC activity or Rewards in general or reducing the rewards of HST Owners

DaBlawb - Flere Ego Te Judico⚡

And this happens way to slow, that’s why we need chainvars to increase PoC activity or Rewards in general or reducing the rewards of HST Owners

Those chain vars have no impact on the actual amount you earn over time. (edited)

3

15:40

What WILL have an impact is approving the HIP and converting all hotspots to light hotspots. Then your earnings will be much more consistent.

3

rawrmaan

The way to increase the reward is by getting more devices on the network that will use data credits and thus cause increased compensation by raising HNT's value. For anyone who has a miner, your rewards in HNT terms will always decrease over time, from now until infinity. That's how the economics of this network have always been (edited)

So you want to tell me the rewards had to halven 4 times from 300000 to 500000 Hotspots? (edited)

DaBlawb - Flere Ego Te Judico⚡

So you want to tell me the rewards had to halven 4 times from 300000 to 500000 Hotspots? (edited)

Active hotspots count, not total on the chain...

Rewards have not moved that much. Your individual hotspot's performance is not indicative of the wider network

rawrmaan

Rewards have not moved that much. Your individual hotspot's performance is not indicative of the wider network

I have 45 Hotspots so I clearly see a pattern

rawrmaan

Rewards have not moved that much. Your individual hotspot's performance is not indicative of the wider network

Well... hang on now. That's not true. The total rewards per day over my fleet have dropped significantly in recent months.

1

It halved exactly 3 times since the actual Halvening in August

ElonTusk | 5Gchipped

As an aside, did anyone get a notification from the app about these votes? I did not.

It just woke me up lol

1

As have the whole network based on the charts (I wish ETL was back up! heh )

1

Looks like some people are complaining more for the rewards there earning than the overall success for Helium. If moving POC to validators is better for the network for substantially and scalability, I'm for it. We currently have more validators than majority of these POS networks.

From 3 HNT to 0.75 HNT a day for my best Hotspot

Guru

Looks like some people are complaining more for the rewards there earning than the overall success for Helium. If moving POC to validators is better for the network for substantially and scalability, I'm for it. We currently have more validators than majority of these POS networks.

One is linked to the other. Let's be real here.

1

15:44

Not saying the move isn't good for the network and good for everyone by extension.

15:45

The upside of this is well worth 0.9%

1

I think nobody will continue to host a Hotspot if the rewards dropped to 1$ a day Better sell it to an idiot that will make 1$ a day (edited)

1

Works out to around, on average, 3 cents a day? Yeah, I'll pay that gladly.

1

ElonTusk | 5Gchipped

One is linked to the other. Let's be real here.

Over 80K of Hotpsots are being added per month to the network with over 3 million on back order. These rewards will always be unsustainable. With or without this move to validators

1

DaBlawb - Flere Ego Te Judico⚡

I think nobody will continue to host a Hotspot if the rewards dropped to 1$ a day Better sell it to an idiot that will make 1$ a day (edited)

OH!! I just looked. Your role is not a validator! lol The role color is so similar!

(edited)

Guru

Over 80K of Hotpsots are being added per month to the network with over 3 million on back order. These rewards will always be unsustainable. With or without this move to validators

Wrong the latest chainvariables decreased it even further by 25-40%

makes so much more sense now

It halved 3 times from 300.000 to 500.000 If it halves 3 times again from 500.000 to 700.000 Helium will ultimately die, take it as a guaranty

rawrmaan

I think it's safe to assume that most validator operators will open the port. And if/when we see that they refuse to do so, we can take it from there. As Abhay said on the call, in this scenario, Helium Inc. will gladly serve the gRPC traffic until a better solution is invented

okay, so there is a plan B in case of problems.

1

I'm curious, are there any validators that will vote "no" on hip-55

Sociotter

I'm curious, are there any validators that will vote "no" on hip-55

Good question.

Sociotter

I'm curious, are there any validators that will vote "no" on hip-55

Do you want to earn even more HNT yes or no? Hard question

1

rawrmaan

Rewards have not moved that much. Your individual hotspot's performance is not indicative of the wider network

Hi, in which HIP we proposed to reduce the max witness from 18 to 14?

1

zee37

Hi, in which HIP we proposed to reduce the max witness from 18 to 14?

This I would like to know as well

Sociotter

I'm curious, are there any validators that will vote "no" on hip-55

Obviously no

That's mathematically false. It reduced 40% in that period. However, it's possible that your hotspot(s) are seeing decreased rewards beyond the mean, and that it's likely not your fault

I still propose this becomes a new service instead of being a validator role... blockchain-challenger and anyone should be able to run one. You have to stake some amount of HNT to be a part of it

1

15:52

Piling everything onto the validators is just going to ultimately have the same end result in scaling

zee37

Hi, in which HIP we proposed to reduce the max witness from 18 to 14?

There wasn't one. That was changed by the Helium core team to increase the reliability for the network. It has no material effect on earnings over time.

1

Wrath - demo.hntmonitor.com

I still propose this becomes a new service instead of being a validator role... blockchain-challenger and anyone should be able to run one. You have to stake some amount of HNT to be a part of it

Interesting idea. You can definitely write a HIP for that

rawrmaan

There wasn't one. That was changed by the Helium core team to increase the reliability for the network. It has no material effect on earnings over time.

You must be joking

rawrmaan

Interesting idea. You can definitely write a HIP for that

wouldnt make it in time for this hip though

DaBlawb - Flere Ego Te Judico⚡

You must be joking

PoC chain variable have never required HIPs and have been changed many times. That's because these variable don't impact the earnings of hotspots, but rather can be tweaked to make the network more stable for everyone. It's a pure positive.

1

Wrath - demo.hntmonitor.com

I still propose this becomes a new service instead of being a validator role... blockchain-challenger and anyone should be able to run one. You have to stake some amount of HNT to be a part of it

Anyone can be a validator...

rawrmaan

There wasn't one. That was changed by the Helium core team to increase the reliability for the network. It has no material effect on earnings over time.

100% of the Hotspots saw the impact on earnings until today, don’t try to fool anyone

3

DaBlawb - Flere Ego Te Judico⚡

100% of the Hotspots saw the impact on earnings until today, don’t try to fool anyone

There are definitely many network issues that are impacting hotspot earnings, but this isn't one of them. It's unrelated

ElonTusk | 5Gchipped

Anyone can be a validator...

with lots of cash though ... that statement is loaded

... ANYONE can run a hotspot, not everyone can be a validator though

1

DaBlawb - Flere Ego Te Judico⚡

100% of the Hotspots saw the impact on earnings until today, don’t try to fool anyone

2

My hotspots have earned the same amount from dec-february, so no 100% have not seen a loss

Wrath - demo.hntmonitor.com

with lots of cash though ... that statement is loaded

... ANYONE can run a hotspot, not everyone can be a validator though

There are pools.

ElonTusk | 5Gchipped

There are pools.

but that doesnt make you a challenger though

Wrath - demo.hntmonitor.com

but that doesnt make you a challenger though

Yes it does.

15:56

Or at least the pool you are in

Ay0hCrypto(ALCRYMIST)

My hotspots have earned the same amount from dec-february, so no 100% have not seen a loss

Yes you are a unique super pony, so 99% saw it

ElonTusk | 5Gchipped

Yes it does.

I see, so you're saying that those rewards would be shared like the cg rewards are?

1

rawrmaan

There wasn't one. That was changed by the Helium core team to increase the reliability for the network. It has no material effect on earnings over time.

One thing is for sure that it reduces our witnessing list which indeed reduced our reward. So, Obviously it has a materialistic effect on our reward as well as on our trust.

1

Wrath - demo.hntmonitor.com

I see, so you're saying that those rewards would be shared like the cg rewards are?

Yes

1

DaBlawb - Flere Ego Te Judico⚡

Yes you are a unique super pony, so 99% saw it

Hard data > your guessing

1

FWIW i havent voted on this yet. lol still debating it

1

ElonTusk | 5Gchipped

Hard data > your guessing

I have 45 Hotspots and I’m monitoring hundreds of Hotspots in Germany

15:59

Your guessing = I don’t care

zee37

One thing is for sure that it reduces our witnessing list which indeed reduced our reward. So, Obviously it has a materialistic effect on our reward as well as on our trust.

It only reduces witnesses per challenge, not total witnesses. You can still obtain the same total amount of unique witnesses over time. Since everyone in the network witnesses less by the same amount, the net effect is zero on any individual hotspot. (edited)

1

DaBlawb - Flere Ego Te Judico⚡

I have 45 Hotspots and I’m monitoring hundreds of Hotspots in Germany

45 vs 540,000+

ElonTusk | 5Gchipped

45 vs 540,000+

So what was the Network average in August after the Halvening? If you talk about facts you better have facts in place

1

rawrmaan

It only reduces witnesses per challenge, not total witnesses. You can still obtain the same total amount of unique witnesses over time. Since everyone in the network witnesses less by the same amount, the net effect is zero on any individual hotspot. (edited)

Just adds to variance for each individual. Small sample sizes for the lose.

DaBlawb - Flere Ego Te Judico⚡

So what was the Network average in August after the Halvening? If you talk about facts you better have facts in place

You know, I might actually have a screenshot still.. hang on...

16:02

2

@DaBlawb - Flere Ego Te Judico⚡ You is a savage haha

rawrmaan

It only reduces witnesses per challenge, not total witnesses. You can still obtain the same total amount of unique witnesses over time. Since everyone in the network witnesses less by the same amount, the net effect is zero on any individual hotspot. (edited)

Over time yes. But for current state of network: No (assuming no change in number of hotspot in certain area)

That would be great, Total amount of Hotspots and average Network rewards to correlate

We weren't even at 100k hotspots total on chain yet

rawrmaan

It only reduces witnesses per challenge, not total witnesses. You can still obtain the same total amount of unique witnesses over time. Since everyone in the network witnesses less by the same amount, the net effect is zero on any individual hotspot. (edited)

Anyway, thanks for your feedback. Have a great day. Enjoy

1

DaBlawb - Flere Ego Te Judico⚡

That would be great, Total amount of Hotspots and average Network rewards to correlate

Sadly until the ETL is back up I can't get current data

(edited)

So we went from 0.59 to 0.14 Does this sound like a healthy way the Network goes? From 300.000 to 500.000??

DaBlawb - Flere Ego Te Judico⚡

So we went from 0.59 to 0.14 Does this sound like a healthy way the Network goes? From 300.000 to 500.000??

300 to 500? What?

So reaching 700.000 Hotspots it will be 0.035 HNT a day on average

16:05

Does this sound like a selfsustaining network growth?

16:06

In 1.5-3 Months

DaBlawb - Flere Ego Te Judico⚡

Does this sound like a selfsustaining network growth?

What does that really have to do with this HIP?

16:06

Also, your math is wrong.

This HIP is part of the problem

1

Those numbers are wrong. 66% growth in hotspots = 66% decrease in earnings per hotspot on average. It's that simple. Your hotspot may see more or less change than that, but is not representative of the total network.

DaBlawb - Flere Ego Te Judico⚡

Do you want to earn even more HNT yes or no? Hard question

I run some validators, not sure if I'm voting yes or no yet. I have similar concerns as PaulM about the HIP as written. This PoC pool is trivial compared to the monthly operating costs, stake, and CG rewards for a validator. My concern is the extra load this puts on CG members might cause perf problems and slow down block production.

1

And yes, this is completely unrelated to this HIP. The only economic change in this HIP is moving 0.9% more rewards to the validators, so they go from 6% to 6.9%

Im extrapolating the reduction from 0.59 to 0.14 HNT a day. What would be your math?

DaBlawb - Flere Ego Te Judico⚡

Im extrapolating the reduction from 0.59 to 0.14 HNT a day. What would be your math?

You're looking at the wrong time range. The graph posted ends when there are 140,000 hotspots on the network. Nobody can grab the latest data right now because the website with the graphs is offline

1

DaBlawb - Flere Ego Te Judico⚡

Im extrapolating the reduction from 0.59 to 0.14 HNT a day. What would be your math?

Each additional hotspot (active) drops the % less than the one before it.

16:09

AKA the slope flattens out

16:10

We've seen large changes in part because hotspots that were on chain, but inactive (not getting rewards) were fixed. You can't look at total hotspots vs rewards.

1

16:10

you have to look at active hotspots.

rawrmaan

Those numbers are wrong. 66% growth in hotspots = 66% decrease in earnings per hotspot on average. It's that simple. Your hotspot may see more or less change than that, but is not representative of the total network.

I know your theoretical fantasy model. But rewards are at 25% from august after Halvening from Network Average 0.59 to 0.14. At the same rate we will be at 0.035 HNT average a Day hitting 750.000 Hotspots. So the coin price needs to 6x in the next 3 Months, otherwise Helium isn’t selfsustaining but instead dieing slowly. And the latest chainvars added to this problem. (edited)

DaBlawb - Flere Ego Te Judico⚡

I know your theoretical fantasy model. But rewards are at 25% from august after Halvening from Network Average 0.59 to 0.14. At the same rate we will be at 0.035 HNT average a Day hitting 750.000 Hotspots. So the coin price needs to 6x in the next 3 Months, otherwise Helium isn’t selfsustaining but instead dieing slowly. And the latest chainvars added to this problem. (edited)

It's not linear...

If it’s not linear it will hit maybe 0.07 HNT a day, the problem stays the same

What would be the onboarding fee for light hotspots?

I'm confused by this being up for vote right now. Validators are struggling with load as it is. They frequently aren't processing quick enough for the network to meet the amount of rewards that are supposed to be done each epoch correct? So wouldn't this just increase their load?

1

1

Jorge DoArt

What would be the onboarding fee for light hotspots?

Same as current hotspots

where are you seeing that "validators are struggling with load"?

1

Helium's own announcements over the past month or two.

cokes

where are you seeing that "validators are struggling with load"?

They think recent issues like the "doom block" were caused by that.

that had nothing to do with validators, but state channel disputes (edited)

2

Helium announcements a while back talked about how validator strain has led to the rewards target not being met on a daily basis.

and how long ago was that?

DaBum

I'm confused by this being up for vote right now. Validators are struggling with load as it is. They frequently aren't processing quick enough for the network to meet the amount of rewards that are supposed to be done each epoch correct? So wouldn't this just increase their load?

This was addressed in the AMA - basically, the validators are already doing most of the work as they have to validate challenge creation anyway

cokes

and how long ago was that?

Few weeks to a month ago I believe

capcom

This was addressed in the AMA - basically, the validators are already doing most of the work as they have to validate challenge creation anyway

So.. there are Validators in here raising concerns about the planned additional load. Can you speak to that concern?

DaBlawb - Flere Ego Te Judico⚡

I know your theoretical fantasy model. But rewards are at 25% from august after Halvening from Network Average 0.59 to 0.14. At the same rate we will be at 0.035 HNT average a Day hitting 750.000 Hotspots. So the coin price needs to 6x in the next 3 Months, otherwise Helium isn’t selfsustaining but instead dieing slowly. And the latest chainvars added to this problem. (edited)

Not really relevant to this HIP, other than to say that some % of rewards volatility is caused by the system that this HIP seeks to replace

1

ElonTusk | 5Gchipped

So.. there are Validators in here raising concerns about the planned additional load. Can you speak to that concern?

To be clear, my concern isn't about the concept of moving PoC generation away from hotspots, I'm concerned about the HIP as written having a hex full of hotspot come check with a CG member to see if they are targeted for PoC. An in-CG validator does have a bunch of time critical work to do and a bunch of new connections can be significant extra work. (edited)

1

zee37

Hi, in which HIP we proposed to reduce the max witness from 18 to 14?

It used to be 5, then it was 25, then it was 18, then it was 14. I’d expect it to change again in the future. None of the changes involved any HIPs

2

@capcom why so long for this hip to come out? Teams been working on light gateways for a long time. One week for the community to understand and vote feels rushed. I totally support light hotspots. There is a lot of nuance in the specific implementation for the community to absorb.

PaulM

@capcom why so long for this hip to come out? Teams been working on light gateways for a long time. One week for the community to understand and vote feels rushed. I totally support light hotspots. There is a lot of nuance in the specific implementation for the community to absorb.

I think we wanted to get more of the implementation finalized before having a HIP to vote on

capcom

I think we wanted to get more of the implementation finalized before having a HIP to vote on

HIP with code? whaaa?

1

Leave that mostly to @hashc0de I wasn’t really involved in the timing or execution

krby

HIP with code? whaaa?

A novel concept!

1

krby

HIP with code? whaaa?

There's also temperature check votes now apparently.

1

Would be nice to talk through the implementation specifics and alternatives longer, though. I couldn't make the AMA this week. 5 days until vote is closed is Tuesday. In there is the weekend plus a likely-for-most US holiday on Monday. (edited)

krby

Would be nice to talk through the implementation specifics and alternatives longer, though. I couldn't make the AMA this week. 5 days until vote is closed is Tuesday. In there is the weekend plus a likely-for-most US holiday on Monday. (edited)

Holiday = All day free to debate and vote!

ElonTusk | 5Gchipped

Holiday = All day free to debate and vote!

I should have been explicit. I don't expect the HIP authors or others on the core team to be avail for debate Sat thru Mon.

krby

I should have been explicit. I don't expect the HIP authors or others on the core team to be avail for debate Sat thru Mon.

ahh gotcha

3

2

3

16:45

my attempted reddit summary

3

Every HIP should have a ELI5 section

capcom

Click to see attachment 🖼️

Wouldn't a firewall still block incoming challenges from a validator?

the requests are outbound from the hotspot, so no

cokes pinned a message to this channel. 02/17/2022 4:47 PM

capcom

the requests are outbound from the hotspot, so no

Ah right.

16:48

goes back to re-read the HIP again

Hip-54 was easy

16:48

Can’t decide what’s happening with hip-55

1

indyitgeek

Can’t decide what’s happening with hip-55

Capcom just gave a really nice summary above

Are we sure? No more sd card issues?

1

quite sure yes

16:50

the only reason they exist today is all the thrashing related to storing the entire blockchain, peerbook, etc

honestly amazing the SD cards last as long as they do with all the I/O they handle

2

Silly question. So the validators are like 10k

1

16:51

Right?

yep

And I don’t own one

capcom

Click to see attachment 🖼️

No more relayed hotspots, just relayed validators, Less SD card issues, Hotspots will be cheaper to make, doesn't mean they will sell for any cheaper. market price is double to triple what distributor price is. doubt that it works out to .9% less rewards, if the top half proves true, it will be more rewards for those with well functioning setups (edited)

indyitgeek

And I don’t own one

You can always join a pool to have a share of one.

And looking at the return on investment

16:52

I should have. Lol

16:52

So how does this change help them

Ay0hCrypto(ALCRYMIST)

No more relayed hotspots, just relayed validators, Less SD card issues, Hotspots will be cheaper to make, doesn't mean they will sell for any cheaper. market price is double to triple what distributor price is. doubt that it works out to .9% less rewards, if the top half proves true, it will be more rewards for those with well functioning setups (edited)

validators are already heavily incentivized to not be relayed, and even if they are, the requests from hotspots come via grpc not libp2p so the entire flow is simpler and more reliable. it's not 9% less, it's 0.9%. that's what you earn today for creating challenges (edited)

Ay0hCrypto(ALCRYMIST)

No more relayed hotspots, just relayed validators, Less SD card issues, Hotspots will be cheaper to make, doesn't mean they will sell for any cheaper. market price is double to triple what distributor price is. doubt that it works out to .9% less rewards, if the top half proves true, it will be more rewards for those with well functioning setups (edited)

*0.9%

I’m not saying I don’t get what your saying

16:53

Ok. I’m cool

1

16:53

Yes on hip-55

capcom

validators are already heavily incentivized to not be relayed, and even if they are, the requests from hotspots come via grpc not libp2p so the entire flow is simpler and more reliable. it's not 9% less, it's 0.9%. that's what you earn today for creating challenges (edited)

typo sorry

1